Obligation Change Detection

Applies formal deontic logic—the logic of obligation and permission—to classify extracted clauses into normative categories: obligations (must), prohibitions (must not), and permissions (may). This structured representation enables the engine to detect when a party's duties shift, even if the surface text is entirely rewritten.

• Maps natural language to modal logic operators
• Distinguishes mandatory duties from aspirational language
• Enables graph-based comparison of normative states

Goes beyond textual redlines to compare the meaning and legal effect of clauses across versions. Uses vector embedding diff techniques to measure cosine distance between obligation-bearing sentences, flagging substantive changes in responsibility that a character-level diff would miss.

• Detects reworded but semantically altered duties
• Identifies shifted risk allocation between parties
• Surfaces implicit obligation changes in defined terms

Constructs a structured network of duties, rights, and conditions extracted from each contract version, then performs a tree edit distance comparison on the resulting obligation graph. This reveals new, removed, or altered normative relationships between contracting parties.

• Nodes represent parties and their obligations
• Edges represent directed duties and conditional triggers
• Graph diff highlights structural changes to the deal

Models time-bound obligations—deadlines, effective dates, renewal windows, and milestone conditions—as first-class entities. The engine detects when temporal constraints are modified, extended, or removed, preventing missed deadlines caused by unnoticed date shifts in dense legal text.

• Extracts and normalizes all date expressions
• Flags changes to performance windows and cure periods
• Generates timeline visualizations of obligation shifts

Tracks changes to capitalized defined terms across contract versions and propagates their modified meanings through every clause that references them. A subtle change to a definition can cascade into sweeping obligation changes—this engine catches that ripple effect automatically.

• Maintains a definition-to-usage index
• Flags semantic drift in key defined terms
• Re-evaluates all dependent clauses on definition change

Provides high-risk analysis that specifically tracks any alteration to the definition or scope of a Material Adverse Change (MAC) clause—a critical condition precedent in M&A transactions. Even minor wording changes to MAC carve-outs can shift billions in deal risk.

• Isolates MAC clause text across versions
• Compares carve-out enumerations and qualifiers
• Generates risk-weighted change severity scores

A structured comparison of the network of duties, rights, and conditions extracted from two contract versions. The system constructs a directed graph where nodes represent parties and edges represent normative relationships. The diff then identifies:

• New obligations imposed on a party
• Removed rights that alter the balance of power
• Modified conditions precedent to performance This graph-level view catches cascading changes that a clause-by-clause comparison would miss.

The algorithmic detection and reconciliation of contradictory legal rules within or across contract versions. When an obligation change introduces a duty that directly conflicts with an existing right—for example, a new exclusivity clause that contradicts a previously granted right to engage subcontractors—the detection engine must flag the normative collision. Advanced systems use defeasible logic to resolve conflicts based on rule priority, specificity, or temporal precedence.

A high-risk analysis that specifically tracks any alteration to the definition or scope of a Material Adverse Change clause. In M&A transactions, the MAC clause is a critical condition precedent that allocates risk between signing and closing. Obligation change detection systems prioritize MAC clause modifications, flagging even subtle changes to carve-outs, thresholds, or look-back periods that could dramatically shift the risk allocation in a deal.